Description

Build and maintain the MLOps platform infrastructure on AWS, managing SageMaker Unified Studio, pipelines, and model serving.

This role is on-site.

Responsibilities

• Configure SageMaker Unified Studio domains, projects, persona-based roles, and multi-environment promotion workflows.
• Build MLOps pipelines using SageMaker Pipelines for data extraction from Snowflake, preprocessing, training, evaluation, and model registration.
• Manage SageMaker Model Registry for cross-account model promotion, versioning, immutability, and lineage tracking.
• Configure MLflow experiment tracking with auto-logging of parameters, metrics, and artifacts.
• Operate model serving via real-time SageMaker endpoints and batch prediction workflows with monitoring for data and model drift.

Required Skills

• 10-15 years of software engineering experience focused on cloud infrastructure or ML platform operations.
• 5+ years hands-on with AWS, including deep expertise in Amazon SageMaker (Studio, Pipelines, Model Registry, Endpoints, Feature Store).
• 3+ years building and operating production MLOps pipelines including training, versioning, deployment, monitoring, and rollback.
• Infrastructure-as-Code with Terraform, CDK, or CloudFormation.
• IAM design for ML platforms including execution roles, service roles, cross-account access, Lake Formation, and SSO/SAML.
• Experience with MLflow or equivalent experiment tracking tools.
• SageMaker Pipelines or similar workflow orchestration (Airflow, Step Functions).
• Model serving expertise: real-time endpoints, batch transform, auto-scaling, and endpoint monitoring.
• Networking and security knowledge: VPC, security groups, private endpoints, and cross-account connectivity.

Preferred Skills

• SageMaker Unified Studio domain provisioning, custom blueprints, and project standardization.
• SageMaker Feature Store for online/offline feature management.
• SageMaker Model Monitor for data quality checks, bias detection, and drift detection.

Key Skills

Education

Bachelor's degree